survinger: Design-Adjusted Inference for Pathogen Lineage Surveillance

Description

Provides tools for optimizing sequencing resource allocation and estimating pathogen lineage prevalence under real-world genomic surveillance conditions. Implements constrained allocation optimization for limited sequencing capacity across multiple regions and sample sources. Includes Horvitz-Thompson and post-stratified estimators that account for unequal sequencing rates, delay-adjusted nowcasting for right-censored reporting data, and combined design-weighted delay-corrected inference with uncertainty propagation.

Author(s)

Maintainer: Cuiwei Gao 48gaocuiwei@gmail.com [copyright holder]

See Also

Useful links:

• https://github.com/CuiweiG/survinger

• Report bugs at https://github.com/CuiweiG/survinger/issues

Assert required columns exist in a data frame

Description

Assert required columns exist in a data frame

Usage

.assert_columns_exist(df, cols, arg_name = "data")

Arguments

Value

Invisible TRUE if all columns exist.

Assert object inherits from a specific surv_* class

Description

Assert object inherits from a specific surv_* class

Usage

.assert_surv_class(x, cls, arg_name = deparse(substitute(x)))

Arguments

Value

Invisible TRUE.

Assert object inherits from surv_design

Description

Assert object inherits from surv_design

Usage

.assert_surv_design(x, arg_name = "design")

Arguments

Value

Invisible TRUE.

Convert Date vector to ISO epiweek strings (YYYY-Wnn)

Description

Convert Date vector to ISO epiweek strings (YYYY-Wnn)

Usage

.date_to_epiweek(dates)

Arguments

Value

Character vector of epiweek labels.

Generate unequal shares summing to 1

Description

Generate unequal shares summing to 1

Usage

.generate_unequal_shares(n, shape = 2)

Arguments

Value

Numeric vector summing to 1.

Kish effective sample size

Description

Kish effective sample size

Usage

.kish_eff_n(weights)

Arguments

Value

Scalar effective sample size.

Create strata key string from data frame rows

Description

Create strata key string from data frame rows

Usage

.make_strata_key(df, strata_vars)

Arguments

Value

Character vector of strata identifiers.

Map stratum-level weights to individual observations

Description

Map stratum-level weights to individual observations

Usage

.map_weights_to_obs(design)

Arguments

Value

Tibble with all data columns plus stratum_id, seq_rate, weight.

Resolve user-specified time column

Description

Handles "epiweek", "month", "date", or a literal column name.

Usage

.resolve_time_column(dat, time, date_col)

Arguments

Value

List with data (possibly modified) and col (column name).

survinger color palette

Description

survinger color palette

Usage

.surv_colors()

Value

A named character vector of hex colors.

Validate and clamp sequencing rates to (0, 1]

Description

Validate and clamp sequencing rates to (0, 1]

Usage

.validate_seq_rates(rates)

Arguments

Value

Numeric vector clamped to 0.001–1.

Wilson score interval for a proportion

Description

Provides valid coverage even at p-hat = 0 or 1, unlike Wald. Reference: Wilson (1927) JASA; Agresti & Coull (1998) TAS.

Usage

.wilson_ci(p, n, z, side = c("lower", "upper"))

Arguments

Value

Scalar bound, clamped to 0–1.

One-row summary of survinger model

Description

One-row summary of survinger model

Usage

## S3 method for class 'surv_prevalence'
glance(x, ...)

## S3 method for class 'surv_delay_fit'
glance(x, ...)

## S3 method for class 'surv_adjusted'
glance(x, ...)

Arguments

Value

A single-row tibble with key summary statistics.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
prev <- surv_lineage_prevalence(d, "BA.2.86")
glance(prev)

Internal constructor for surv_allocation

Description

Internal constructor for surv_allocation

Usage

new_surv_allocation(
  allocation,
  objective,
  total_capacity,
  constraints,
  diagnostics
)

Internal constructor for surv_delay_fit

Description

Internal constructor for surv_delay_fit

Usage

new_surv_delay_fit(distribution, parameters, strata, data_summary, diagnostics)

Internal constructor for surv_prevalence

Description

Internal constructor for surv_prevalence

Usage

new_surv_prevalence(estimates, design, method, lineage, conf_level, time_unit)

Plot methods for survinger objects

Description

Visualize surveillance design, allocation, prevalence estimates, delay distributions, nowcasts, and adjusted estimates.

Usage

## S3 method for class 'surv_design'
plot(x, ...)

## S3 method for class 'surv_allocation'
plot(x, ...)

## S3 method for class 'surv_prevalence'
plot(x, ...)

## S3 method for class 'surv_delay_fit'
plot(x, ...)

## S3 method for class 'surv_nowcast'
plot(x, ...)

## S3 method for class 'surv_adjusted'
plot(x, ...)

Arguments

Value

A ggplot2 object.

Combined design-weighted and delay-adjusted prevalence

Description

Simultaneously corrects for unequal sequencing rates and right-truncation from reporting delays.

Usage

## S3 method for class 'surv_adjusted'
print(x, ...)

## S3 method for class 'surv_adjusted'
as.data.frame(x, ...)

surv_adjusted_prevalence(
  design,
  delay_fit,
  lineage,
  time = "epiweek",
  prevalence_method = "hajek",
  nowcast_method = "direct",
  conf_level = 0.95,
  bootstrap_n = 0L
)

Arguments

Value

Invisibly returns the input object.

A surv_adjusted object.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 12, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
delay <- surv_estimate_delay(d)
adj <- surv_adjusted_prevalence(d, delay, "BA.2.86")
print(adj)

Optimize sequencing allocation across strata

Description

Given fixed total sequencing capacity, finds the optimal allocation across strata that minimizes a specified objective function.

Usage

## S3 method for class 'surv_allocation'
print(x, ...)

## S3 method for class 'surv_allocation'
as.data.frame(x, ...)

surv_optimize_allocation(
  design,
  objective = c("min_mse", "max_detection", "min_imbalance"),
  total_capacity,
  budget = NULL,
  min_per_stratum = 2L,
  target_lineage = NULL,
  target_prevalence = 0.01,
  cost_col = NULL
)

Arguments

Value

Invisibly returns the input object.

A surv_allocation object.

Examples

sim <- surv_simulate(n_regions = 4, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
a <- surv_optimize_allocation(d, "min_mse", total_capacity = 500)
print(a)

Estimate reporting delay distribution

Description

Fits a parametric distribution to the delay between sample collection and sequence reporting, accounting for right-truncation.

Usage

## S3 method for class 'surv_delay_fit'
print(x, ...)

surv_estimate_delay(
  design,
  distribution = c("negbin", "poisson", "lognormal", "nonparametric"),
  strata = NULL,
  max_delay = 60L,
  ref_date = NULL
)

Arguments

Value

Invisibly returns the input object.

A surv_delay_fit object.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 12, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
fit <- surv_estimate_delay(d)
print(fit)

Create a genomic surveillance design object

Description

Constructs a survey design object tailored for pathogen genomic surveillance, encoding stratification structure, sequencing rates, and population information needed for design-weighted inference.

Usage

## S3 method for class 'surv_design'
print(x, ...)

## S3 method for class 'surv_design'
summary(object, ...)

## S3 method for class 'summary.surv_design'
print(x, ...)

surv_design(
  data,
  strata,
  sequencing_rate,
  population,
  date_collected = "collection_date",
  date_reported = "report_date",
  lineage = "lineage",
  source_type = NULL,
  source_config = NULL
)

Arguments

Value

Invisibly returns the input object.

A summary list of class summary.surv_design.

Invisibly returns the input object.

An object of class surv_design.

See Also

surv_simulate(), surv_lineage_prevalence(), surv_optimize_allocation()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 8, seed = 42)
design <- surv_design(
  data = sim$sequences,
  strata = ~ region,
  sequencing_rate = sim$population[c("region", "seq_rate")],
  population = sim$population
)
print(design)

Nowcast lineage counts correcting for reporting delays

Description

Nowcast lineage counts correcting for reporting delays

Usage

## S3 method for class 'surv_nowcast'
print(x, ...)

## S3 method for class 'surv_nowcast'
as.data.frame(x, ...)

surv_nowcast_lineage(
  design,
  delay_fit,
  lineage = NULL,
  time = "epiweek",
  horizon = 4L,
  ref_date = NULL,
  method = c("direct", "em")
)

Arguments

Value

Invisibly returns the input object.

A surv_nowcast object.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 12, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
fit <- surv_estimate_delay(d)
nc <- surv_nowcast_lineage(d, fit, "BA.2.86")
print(nc)

Estimate lineage prevalence with design weights

Description

Estimates the prevalence of a specified pathogen lineage over time, correcting for unequal sequencing rates across strata.

Usage

## S3 method for class 'surv_prevalence'
print(x, ...)

## S3 method for class 'surv_prevalence'
as.data.frame(x, ...)

surv_lineage_prevalence(
  design,
  lineage,
  time = "epiweek",
  method = c("hajek", "horvitz_thompson", "poststratified"),
  conf_level = 0.95,
  min_obs = 5L
)

Arguments

Value

Invisibly returns the input object.

A data.frame.

A surv_prevalence object.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
prev <- surv_lineage_prevalence(d, "BA.2.86")
print(prev)

Objects exported from other packages

Description

These objects are imported from other packages. Follow the links below to see their documentation.

generics

glance, tidy

Example SARS-CoV-2 genomic surveillance data

Description

Simulated genomic surveillance dataset with 5 regions, 26 weeks, highly unequal sequencing rates (15% to 0.5%), three sample sources, and negative binomial reporting delays. Contains known ground truth for benchmarking.

Usage

sarscov2_surveillance

Format

A named list with four elements:

sequences

Tibble of sequence records: sequence_id, region, source_type, lineage, collection_date, report_date, epiweek.

population

Tibble with one row per region: region, n_positive, n_sequenced, seq_rate, pop_total.

truth

Tibble of true lineage prevalence by region and week.

parameters

List of simulation parameters.

Source

Simulated using surv_simulate(seed = 20240101).

Examples

data(sarscov2_surveillance)
head(sarscov2_surveillance$sequences)
sarscov2_surveillance$population

Combine multiple prevalence estimates

Description

Bind results from different lineages or methods into a single tibble for comparison plots and tables.

Usage

surv_bind(...)

Arguments

Value

A tibble with a source column identifying each input.

See Also

surv_lineage_prevalence(), surv_sensitivity()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
p1 <- surv_lineage_prevalence(d, "BA.5")
p2 <- surv_lineage_prevalence(d, "XBB.1.5")
combined <- surv_bind(p1, p2)
head(combined)

Compare multiple allocation strategies

Description

Compare multiple allocation strategies

Usage

surv_compare_allocations(
  design,
  strategies = c("equal", "proportional", "min_mse", "max_detection", "min_imbalance"),
  total_capacity,
  target_prevalence = 0.01,
  ...
)

Arguments

Value

A tibble comparing strategies.

Examples

sim <- surv_simulate(n_regions = 4, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
surv_compare_allocations(d, total_capacity = 200)

Compare weighted vs naive prevalence estimates

Description

Side-by-side plot showing the impact of design correction.

Usage

surv_compare_estimates(weighted, naive, title = NULL)

Arguments

Value

A ggplot2 object.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
w <- surv_lineage_prevalence(d, "BA.2.86")
n <- surv_naive_prevalence(d, "BA.2.86")
surv_compare_estimates(w, n)

Compute design effect over time

Description

Compute design effect over time

Usage

surv_design_effect(weighted, naive)

Arguments

Value

A tibble with time, deff, and bias_correction columns.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
w <- surv_lineage_prevalence(d, "BA.2.86")
n <- surv_naive_prevalence(d, "BA.2.86")
surv_design_effect(w, n)

Variant detection probability under current design

Description

Variant detection probability under current design

Usage

surv_detection_probability(
  design,
  true_prevalence,
  delay_fit = NULL,
  n_periods = 1L,
  detection_threshold = 1L
)

Arguments

Value

A list with overall, cumulative, by_stratum, parameters.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
surv_detection_probability(d, 0.01)

Pipe-friendly surveillance analysis

Description

Convenience wrapper that creates a design, estimates prevalence, and optionally applies delay correction in a single pipe-friendly call. Designed for rapid exploratory analysis in interactive sessions.

Usage

surv_estimate(
  data,
  strata,
  sequencing_rate,
  population,
  lineage,
  correct_delay = FALSE,
  method = "hajek",
  ...
)

Arguments

Value

A surv_prevalence or surv_adjusted object.

See Also

surv_design(), surv_lineage_prevalence()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
# One-liner analysis:
result <- surv_estimate(
  data = sim$sequences, strata = ~ region,
  sequencing_rate = sim$population[c("region", "seq_rate")],
  population = sim$population,
  lineage = "BA.2.86"
)
print(result)

Subset a surveillance design by filter criteria

Description

Creates a new surv_design object containing only sequences matching the specified filter criteria.

Usage

surv_filter(design, ...)

Arguments

Value

A new surv_design object with filtered data.

See Also

surv_design()

Examples

sim <- surv_simulate(n_regions = 5, n_weeks = 12, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
d_sub <- surv_filter(d, region %in% c("Region_A", "Region_B"))
print(d_sub)

Compute naive (unweighted) lineage prevalence

Description

Simple proportion without design correction. Useful as baseline.

Usage

surv_naive_prevalence(design, lineage, time = "epiweek", conf_level = 0.95)

Arguments

Value

A surv_prevalence object with method = "naive".

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
naive <- surv_naive_prevalence(d, "BA.2.86")

Plot allocation plan

Description

Plot allocation plan

Usage

surv_plot_allocation(allocation)

Arguments

Value

A ggplot2 object.

Examples

sim <- surv_simulate(n_regions = 5, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
a <- surv_optimize_allocation(d, "min_mse", total_capacity = 500)
surv_plot_allocation(a)

Plot sequencing rate inequality across strata

Description

Plot sequencing rate inequality across strata

Usage

surv_plot_sequencing_rates(design)

Arguments

Value

A ggplot2 object.

Examples

sim <- surv_simulate(n_regions = 5, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
surv_plot_sequencing_rates(d)

Compute power curve for detection across prevalence range

Description

Generates a detection probability curve that can be directly plotted or included in publications. Answers: "At what prevalence does our surveillance achieve X% detection?"

Usage

surv_power_curve(
  design,
  prevalence_range = seq(0.001, 0.05, by = 0.001),
  delay_fit = NULL,
  thresholds = c(0.5, 0.8, 0.95)
)

## S3 method for class 'surv_power_curve'
plot(x, ...)

Arguments

Value

A list with:

curve

Tibble with prevalence and detection columns.

thresholds

Tibble with threshold, prevalence_needed columns.

A ggplot2 object.

See Also

surv_detection_probability(), surv_required_sequences()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
pc <- surv_power_curve(d)
pc$thresholds

Estimate prevalence by subgroup

Description

Applies surv_lineage_prevalence() within subgroups defined by a grouping variable. Analogous to survey::svyby() for stratified survey analysis.

Usage

surv_prevalence_by(
  design,
  lineage,
  by,
  time = "epiweek",
  method = "hajek",
  conf_level = 0.95
)

Arguments

Value

A tibble with columns: group, time, lineage, prevalence, se, ci_lower, ci_upper, n_obs, effective_n.

See Also

surv_lineage_prevalence(), surv_filter()

Examples

sim <- surv_simulate(n_regions = 4, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")],
                 sim$population, source_type = "source_type")
surv_prevalence_by(d, "BA.2.86", by = "region")

Compute surveillance quality metrics

Description

Returns a single-row tibble of design quality indicators suitable for inclusion in manuscripts. Analogous to broom::glance() but for the surveillance design itself.

Usage

surv_quality(design, target_lineage = NULL, target_prevalence = 0.01)

Arguments

Value

A single-row tibble with columns: n_obs, n_strata, gini, rate_ratio, effective_n, deff, detection_prob, mean_bias.

See Also

surv_report(), surv_design()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
surv_quality(d)

Generate a comprehensive surveillance system report

Description

Produces a summary of the current surveillance design's strengths, weaknesses, and recommendations.

Usage

surv_report(design, target_lineage = NULL, target_prevalence = 0.01)

Arguments

Value

Invisibly returns a named list of computed metrics including n_obs, n_strata, rate_range, gini, effective_n, detection_prob, and mean_bias.

See Also

surv_design(), surv_lineage_prevalence(), surv_detection_probability(), surv_optimize_allocation()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
surv_report(d)

Compute cumulative reporting probability

Description

Compute cumulative reporting probability

Usage

surv_reporting_probability(delay_fit, delta, stratum = NULL)

Arguments

Value

Numeric vector of probabilities.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 12, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
fit <- surv_estimate_delay(d)
surv_reporting_probability(fit, delta = c(7, 14, 21))

Required sequences for target detection probability

Description

Required sequences for target detection probability

Usage

surv_required_sequences(
  true_prevalence,
  target_probability = 0.95,
  n_periods = 1L,
  detection_threshold = 1L
)

Arguments

Value

Integer.

Examples

surv_required_sequences(0.01)
surv_required_sequences(0.05, target_probability = 0.99)

Sensitivity analysis across methods

Description

Runs all three prevalence estimators and delay/no-delay variants on the same design, producing a comparison table. Essential for robustness checks in publications.

Usage

surv_sensitivity(
  design,
  lineage,
  delay_fit = NULL,
  time = "epiweek",
  conf_level = 0.95
)

Arguments

Value

A tibble with one row per method-time combination, columns: method, time, prevalence, se, ci_lower, ci_upper.

See Also

surv_lineage_prevalence(), surv_adjusted_prevalence()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
surv_sensitivity(d, "BA.2.86")

Override design weights with custom values

Description

Override design weights with custom values

Usage

surv_set_weights(design, weights)

Arguments

Value

Updated surv_design.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 4, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
d2 <- surv_set_weights(d, rep(1.0, d$n_strata))

Simulate genomic surveillance data

Description

Generates synthetic surveillance datasets with realistic features: multiple regions with unequal sequencing rates, multiple lineages with time-varying prevalence, configurable reporting delays, and multiple sample sources.

Usage

surv_simulate(
  n_regions = 5L,
  n_weeks = 26L,
  total_positive_per_week = 1000L,
  sequencing_rates = NULL,
  lineage_dynamics = NULL,
  delay_params = list(mu = 10, size = 3),
  sources = c("clinical", "wastewater", "sentinel"),
  source_weights = c(0.7, 0.2, 0.1),
  seed = NULL
)

Arguments

Value

A named list with elements:

sequences

Tibble of individual sequence records.

population

Tibble with one row per region.

truth

Tibble of true lineage prevalence by region and week.

parameters

List of all input parameters.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 8, seed = 42)
head(sim$sequences)
sim$population

Format prevalence results for knitr tables

Description

Produces a publication-ready summary table from any survinger result. Automatically called when objects are printed inside RMarkdown chunks if knitr is loaded.

Usage

surv_table(x, digits = 3, percent = TRUE, ...)

Arguments

Value

A tibble formatted for display.

See Also

surv_lineage_prevalence(), tidy.surv_prevalence()

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
prev <- surv_lineage_prevalence(d, "BA.2.86")
surv_table(prev)

Update sequencing rates in a surveillance design

Description

Update sequencing rates in a surveillance design

Usage

surv_update_rates(design, new_rates)

Arguments

Value

Updated surv_design.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 4, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
new_r <- sim$population[c("region", "seq_rate")]
new_r$seq_rate <- new_r$seq_rate * 2
d2 <- surv_update_rates(d, new_r)

Publication-quality ggplot2 theme

Description

A clean, high-contrast theme suitable for journal publication. Follows Nature/Lancet figure guidelines.

Usage

theme_survinger(base_size = 11, base_family = "")

Arguments

Value

A ggplot2 theme object.

See Also

surv_compare_estimates()

Examples

library(ggplot2)
ggplot(mtcars, aes(wt, mpg)) + geom_point() + theme_survinger()

Extract tidy estimates from survinger objects

Description

Converts survinger result objects into tidy tibbles suitable for further analysis with dplyr, ggplot2, or other tidyverse tools.

Usage

## S3 method for class 'surv_prevalence'
tidy(x, ...)

## S3 method for class 'surv_nowcast'
tidy(x, ...)

## S3 method for class 'surv_adjusted'
tidy(x, ...)

## S3 method for class 'surv_allocation'
tidy(x, ...)

## S3 method for class 'surv_delay_fit'
tidy(x, ...)

Arguments

Value

A tibble.

Examples

sim <- surv_simulate(n_regions = 3, n_weeks = 10, seed = 1)
d <- surv_design(sim$sequences, ~ region,
                 sim$population[c("region", "seq_rate")], sim$population)
prev <- surv_lineage_prevalence(d, "BA.2.86")
tidy(prev)